Okurigana are the kana written after a kanji to complete a word: the る in 食べる, the しい in 新しい. They exist because Japanese inflects and kanji do not, and getting them right is a writing skill that recognition practice never touches.
What they do
A kanji carries a meaning and cannot show grammar. 食 is about eating; whether the word is eat, ate, eating or does not eat is carried by the kana that follow.
So 食べる taberu, to eat, becomes 食べた tabeta, ate, and 食べない tabenai, does not eat. The kanji stays; the kana change.
They also disambiguate. The same character can begin several different words, and the okurigana signal which one: 上がる agaru and 上げる ageru share a kanji and differ in the kana, with different meanings.
That disambiguating role is why okurigana conventions are standardised rather than left to preference. Japan’s jōyō kanji table and its accompanying guidance set out the conventions for how much of a word is written in kana.
| Written form | Reading | Meaning |
|---|---|---|
| 食べる | taberu | to eat |
| 食べた | tabeta | ate |
| 上がる | agaru | to go up |
| 上げる | ageru | to raise |
| 新しい | atarashii | new |
Why learners get them wrong
Three failure modes, all of them specific to writing rather than reading.
Writing too little. 食る instead of 食べる, because the learner remembers the kanji and not where the word’s kana begin. Reading practice never surfaces this, since the correct form is always on the page.
Writing too much. 新らしい instead of 新しい, usually from over-applying a pattern.
Choosing the wrong pair. 上がる where 上げる is meant, which changes the meaning rather than just the spelling, and which a reader notices immediately.
All three come from learning the character rather than the word, which is why practising vocabulary rather than isolated kanji prevents them almost entirely.
The patterns that help
Okurigana are not arbitrary, and a few observations cover most cases.
Verb endings are written in kana from the point where the word inflects. That is why 食べる has べる and 見る has る: the kanji covers the unchanging part and the kana carry what changes.
Adjective endings follow the same logic. 新しい keeps しい in kana because that is the inflecting portion.
Pairs of related verbs are distinguished by their okurigana, and learning them as a pair rather than separately is far more efficient: 上がる and 上げる, 出る and 出す, 入る and 入れる.
And nouns generally have no okurigana at all, which is why 食事 and 新聞 are written in kanji alone.
Why this matters for handwriting specifically
When you type, the input method produces the whole word including its kana, so the okurigana are supplied. When you write by hand, they are yours to produce, and they are as much part of the word as the kanji.
This is the clearest argument for practising words rather than characters. A learner drilling 食 alone will write 食 correctly and still fail to write 食べる, and the failure is invisible until the moment it matters.
It also affects appearance. Kana are generally smaller and rounder than the kanji beside them, and the okurigana should read as an ending rather than as a separate word, which means slightly tighter spacing. Practising kanji in isolation never rehearses either.
Research suggests the production practice is worth it beyond correctness: a brain-wave study found words learned by handwriting were encoded more strongly than the same words typed, and okurigana are precisely the part of a word that typing hides.
Where okurigana and typing diverge
The reason okurigana errors surprise people is that typing conceals them completely.
Typing たべる produces 食べる with the kana already in place. You never decide where the kanji ends and the kana begin, because the input method decided.
Handwriting removes that support entirely, and it is the only common situation that does. Reading shows you the answer; typing supplies it; only writing asks you to produce it.
That makes okurigana a good diagnostic for whether your practice is genuinely production-based. If you have never gotten one wrong, either you have been practising words by hand for a while or you have not been testing production at all.
How to practise them
Prompt with the word, not the character. “taberu, to eat” rather than “eat”, and write 食べる in full, kanji and kana together.
Check the whole word. A correct kanji with wrong okurigana is a wrong word, and grading it as correct trains the error.
Drill the confusable pairs deliberately. 上がる against 上げる, 出る against 出す. They fail as pairs, so practising them apart leaves the confusion intact.
Write short sentences weekly. Okurigana sit in a stream of kana and kanji, and the spacing and proportion only come right in context.
| Practice item | Trains | What it misses |
|---|---|---|
| Kanji alone | The character | The word, the okurigana |
| Word with okurigana | The written word | Sentence-level spacing |
| Confusable pair together | The distinction | Nothing; do this deliberately |
| Short sentence | Everything in context | Slower, do it weekly |
Where okurigana affect meaning most
Some pairs are worth learning deliberately because getting them wrong changes what you said rather than how it looks.
Transitive and intransitive verb pairs are the largest group: 上がる and 上げる, 出る and 出す, 入る and 入れる, 開く and 開ける. One describes something happening, the other describes someone doing it, and the kanji is identical.
Words that share a kanji and diverge entirely are the second group, where the okurigana are the only signal about which word is meant.
And adjective forms, where the inflecting portion carries tense and negation, so writing too little produces a form that is simply not a word.
The efficient response is to learn these as pairs from the beginning rather than as separate items. A pair practised together teaches the distinction; the same two learned months apart leaves the confusion in place and takes longer to fix later.
The dictionary convention
When you look up a verb, dictionaries show the plain form with its okurigana, and that is the form worth learning first, because every other form is derived from it. The jōyō kanji table and its accompanying guidance are the reference for which forms are standard in public writing.
Some words have accepted variant okurigana, where more or less of the word may be written in kana, and both forms appear in real writing. The official guidance sets a standard for public documents; ordinary writing tolerates variation.
The practical position for a learner: learn the standard form, recognise that variants exist, and do not spend energy on edge cases until you meet one in something you are reading.
{{appName}} prompts with the whole word so the okurigana are part of what you produce, keeps the characters hidden until you have written them, and checks stroke order on your attempt. It is free in early access. For the wider method, kanji compounds practice and how to practise writing kanji are the companion pieces.
The reading side, briefly
Okurigana also do a job for readers, which is worth understanding because it explains why the conventions are stable.
Japanese is written without spaces, so the reader has to work out where words begin and end. Kanji mark content words and kana mark grammar, and that alternation is a large part of what makes the text parseable at a glance.
Okurigana sit exactly on that boundary. Seeing 食べ tells a reader that a verb starts here and is inflecting, before they have processed the rest of the word.
That is why writing too little is a genuine reading problem rather than a spelling preference: 食る removes the signal the reader uses, and the sentence becomes momentarily ambiguous.
Understanding this also makes the conventions easier to remember, since they stop being arbitrary rules and become a system for marking word boundaries in a script that has no spaces.
Practising the pairs that matter
A short deliberate drill covers most okurigana errors, and it takes about ten minutes once.
List the transitive and intransitive pairs you have already met: 上がる and 上げる, 出る and 出す, 入る and 入れる, 開く and 開ける, 始まる and 始める.
Write each pair together, from a prompt, with the meanings stated: something goes up against someone raises it. Writing them side by side is what teaches the distinction, since the confusion lives in the pair rather than in either word.
Then use each in a short sentence, because the difference is grammatical and only a sentence exercises it.
Repeat the list once a week for a month and the pairs stop being a source of error. Doing it by hand matters here as elsewhere: a brain-wave study found words learned by handwriting were encoded more strongly than the same words typed, and these particular words are ones typing supplies for you every time.
Variation, and how much to worry about it
Some words accept more than one okurigana pattern, which unsettles learners more than it should.
Official guidance sets a standard for public documents, and ordinary writing tolerates variation, particularly in words where a shorter or longer kana portion has become customary in a particular field or publication.
The practical rule: learn the standard form your material uses, recognise the variant if you meet it, and do not spend study time on which is more correct. Both will be read without hesitation by a native reader.
The one place to be strict is the pairs where okurigana carry meaning rather than convention. Those are not variation, they are different words, and the distinction is worth being precise about.
Frequently asked questions
What is okurigana?
The kana written after a kanji to complete a word, such as the べる in 食べる or the しい in 新しい. They exist because Japanese inflects while kanji cannot show grammar, so the character carries the unchanging meaning and the kana carry tense, negation and other changes. They also distinguish related words that share a kanji.
Why do I keep getting okurigana wrong?
Almost always because you learned the character rather than the word. Reading never surfaces the problem, since the correct form is always printed, and typing hides it, since the input method supplies the kana. Practising with the whole word as the prompt, and checking the whole word when grading, prevents nearly all of it.
Are there rules for okurigana?
There are conventions rather than absolute rules, standardised in official guidance for public documents. The useful patterns: verb and adjective endings are written in kana from the point the word inflects, related verb pairs are distinguished by their okurigana, and nouns generally have none. Some words have accepted variants where more or less is written in kana.
How do I practise okurigana?
Prompt yourself with the word and its reading rather than the character, write the whole thing including the kana, and check the whole word rather than just the kanji. Drill confusable pairs such as 上がる and 上げる together, since they fail as pairs. Then write short sentences weekly so spacing and proportion come right in context.
What is the best app for practising whole words rather than characters?
{{appName}} prompts with the word so the okurigana are part of what you produce, keeps the characters hidden until you have written them, checks the stroke order of your attempt and returns the ones you miss sooner. Practising words is what prevents the specific failure of a correct kanji with wrong kana. It is free in early access.
